The SeniorCare Investor: CNL Retirement Properties And The Role Of Capital (con't)
Most of the dollars invested by CNL in 2003 were related to only four deals with a combined purchase price of just over $700 million. Two of the four transactions involved former Marriott Senior Living (MSL) properties now managed by Sunrise Senior Living (NYSE: SRZ); one transaction involved SRZ facilities with Sunrise staying on to manage them; and the fourth was the purchase of the EdenCare Senior Living properties, with SRZ hired to manage all but three of them.
The first of the four, the purchase of nine MSL facilities for $167.6 million, or $86,600 per unit, appears to be the cheapest based on traditional valuation methods. The price was below replacement cost, as seven of the nine were built in the past four years. The remaining two, both built in 1989, are large CCRCs that contain a combined 1,003 units (51% of the total units in the package), but have average occupancy rates above 96%. The newer facilities have struggled, with occupancy rates ranging from 45.2% to 85.1% in 2003, with an average of just over 73%. Based on the first quarter 2003 annualized financial data, the cap rate was 10.2%, the price to revenue multiple was 2.0x and EBITDA covers the first year lease payment by a small margin. This seems very reasonable, especially if one believes that Sunrise will increase the occupancy levels of the newer facilities, with most of that cash flow going to the bottom line. And with the two CCRCs, the actual cash flow from the turnover of the independent living apartments is higher than the EBITDA presented. So far, so good.
The next deal, which closed on August 29 last year, involved the purchase of 14 MSL facilities for $184.5 million, or $115,100 per unit. All of these facilities opened in 1998 or 1999, and the average occupancy for the first nine months of 2003 was just under 80%. One of the problems, however, is that the occupancy trend had declined at seven of the facilities when compared with 2002. Based on annualized adjusted EBITDA for the first six months of 2003, we calculated an 8.4% cap rate and a price to revenue multiple of 2.8x. The annualized EBITDA falls short of the first year rent by more than $2.0 million, but we were unable to take into account the increased occupancy assumptions. If it takes two years to reach 90% occupancy for the group, at an average current daily rent of $125 per occupied unit, the lease payment will be covered. We do not know if anyone (Marriott or Sunrise) is guaranteeing the near-term shortfall.
The third transaction, which closed on September 30, gets a little dicier. The price of $158.5 million for 16 communities comes to just over $150,000 per unit, but the cap rate based on an adjusted annualized EBITDA for the first nine months of 2003 is 7.8% and the price to revenue multiple is 3.6x. These were already operated by Sunrise and are a mix of old (10 years) and new, with an average occupancy rate for the first nine months of 2003 of 83%. The average daily rent per occupied unit of $149 is higher than the MSL properties, which helps explain how the adjusted EBITDA just covered the 2003 lease payment but falls short by $2.5 million in 2004.
Obviously, this does not take into account any increase in occupancy, but it appears that Sunrise has been struggling with these facilities. Now this is a case where one may wonder what CNL was thinking. But a closer look at the documents shows that Sunrise (as manager only) has guaranteed, for a minimum of 30 months, any shortfall in lease payments and required FFE reserves. Consequently, CNL has breathing room of at least two and one-half years, and expects that occupancy will rise and rates will increase by at least inflation to cover future lease payments. This was an aspect of the deal that did not come to light until recently, and makes the transaction more understandable from a price and risk perspective.
And then came EdenCare Senior Living. It had been well known in the industry that EdenCare, based in Georgia, had been struggling with occupancy issues. Excluding three facilities in Florida that were purchased by CNL but that Sunrise will not be managing (which had average 2003 occupancy rates in excess of 93%), the remaining 22 facilities had an average 76% occupancy in the first nine months of 2003, and a little more than three-quarters of these opened in 2000 or earlier. The price for all 25 was $198.7 million, or $103,900 per unit, which is most likely below replacement cost. The cap rate on the adjusted annualized EBITDA for the first nine months of 2003 (for all 25 properties) was 5.4%, and the price to revenue multiple was 3.7x. Unless occupancy increases soon, the first year lease payment of $19.5 million far exceeds the 2003 cash flow.
Now, CNL did not underwrite based on these financials, because even its cost of capital is above 5.7%. Once again, the company believes that under Sunrise’s management, the occupancy levels will increase, and CNL does not plan to take the hit while they do rise. As part of the deal, EdenCare had to put $5.5 million into escrow to cover any lease and FFE funding shortfalls. On top of this, Sunrise has guaranteed an additional $5.0 million, to be drawn down only when the initial $5.5 million has been used up. When the transaction was first announced, we had to make the assumption that someone was providing guarantees, but at the time we did not know who. It now looks like some of Sunrise’s "risk-free" management contracts carry a bit of risk after all.
There should be a common theme running through these CNL transactions by now. Not one of the purchases was made based on historical financial performance, which certainly would not have justified the pricing in at least three of the four deals. CNL has basically paid a "forward price," what it believes the facilities will be worth in one to two years, and has structured the financial terms in most cases to protect itself until an assumed stabilization period. CNL management is extremely positive on the long-term prospects of the senior care industry, and believes that because there is little new development on a national basis, once the current supply is absorbed, demand will exceed supply and rental rates will then have the potential to increase at a rate greater than inflation.
Is this a risky way to invest in the industry? Yes, because of the uncertainty, and other buyers (operators) would discount that risk in the price. But the risk is mitigated to a degree by the stop-gap guarantees mentioned, but also by CNL’s capital structure. The company’s debt to asset ratio is only 24% today, with a target of 40% to 50%, which may or may not be reached. CNL has been assuming some attractive debt in its deals, and refinancing other debt at relatively low rates (lower than its current dividend yield). So if the projections fall short by a year or two, solvency is not an issue and the equity investors (you remember, the ones without any teeth) could see their yield drop temporarily, but there won’t be much they can do about it. Time, it seems, will be on CNL’s side, and it will have the luxury of waiting until 2008 if it has to.
The question that everyone is asking, however, is whether that is enough time to provide the current return which CNL’s limited partners are expecting plus the full repayment of the original principal invested. Between sales commissions and other fees, a $10,000 investment quickly becomes $8,500 to $9,000 available for actual investment in the senior care properties CNL identifies. When the acquisition prices are a bit rich to start with and based on assumptions that may not materialize, trying to recoup up to an additional 15% of the original invested capital may be a tall order. The fear, of course, is that if investors do not get their original investment back, the funds will dry up and CNL will be left nursing its wounds. For better or worse, we will not have an answer to this issue for several years, but past history tells us that these kinds of high front-end load investment vehicles often don’t fulfill their promises (even though no promises are technically made).
The Horizon Bay Deal
CNL did not waste any time this year, announcing its largest deal to date with the $562 million acquisition of the 20 communities managed by Horizon Bay Senior Communities and owned by WHSLH Realty, L.L.C. and WHSLC Realty, L.L.C., both of which are affiliates of Goldman Sachs. The price came in at $156,500 per unit, and the portfolio is comprised of 2,649 independent living, 716 assisted living and 66 Alzheimer’s units, and four of the communities have a total of 159 skilled nursing beds. Only two of the facilities have fewer than 100 units/beds, while one-third of the communities have over 200 units each. This was a big price, surprising many people in the industry and obviously pleasing the owners of the real estate, which included an affiliate of Chicago-based Senior Lifestyle Corp.
The biggest surprise, and the one causing the most consternation in the market, is the theoretical cap rate for the deal. Based on the annualized figures for the first nine months of 2003, the cap rate was a low 6.0%, and the price to revenue multiple was a high 5.35x. Once again, CNL did not underwrite the acquisition assuming that the 20 properties would throw off just $34 million of cash flow, especially since the first year’s lease payments total $43.4 million, rising to $48.2 million in the second year. The rumors in the market are that the cap rate was somewhere between 8% and 9% based on 2005 cash flow. This would seem to make sense since the lease yield in 2005 is 8.5%, and presumably by then the cash flow would have to start covering lease payments. Unfortunately, CNL management is prohibited from making comments on future performance.
The market’s reaction has been that CNL was, basically, nuts to offer this kind of price, and that Goldman is laughing all the way to the bank. We shared those views as well, especially since a substantial increase in EBITDA is necessary to reach the lease payments required (at least based on the cash flow for the first nine months of 2003). Although we do not know the occupancy of the portfolio, we do know that Horizon Bay just spent, on average, more than $10,000 per unit in renovating its portfolio. Our guess is that the funds were concentrated on select communities and that this was done to increase occupancy, especially in some of the communities that may have needed to be repositioned in their markets. With this work now completed, occupancy levels can be expected to rise from 2003 levels, as well as the monthly rates, with most of the additional income going straight to the bottom line.
Our estimate is that each one percentage point increase in occupancy will produce close to $1.0 million of EBITDA, and each one percentage point increase in the monthly rate produces more than $1.0 million in incremental revenue (which is on top of the occupancy increase). We give the benefit of the doubt to CNL that this is what they were looking at.
WHSLC Realty or one of its affiliates will be leasing the properties from CNL, with Horizon Bay Management, 100% owned by WHSLH Realty, continuing to manage the 20 properties. Because of the continued involvement of the Goldman Sachs affiliate, and knowing that CNL has structured these deals, where the apparent cash flow does not cover lease payments, with initial period guarantees or escrow deposits to cover shortfalls, we have to assume that this arrangement exists here. Management can’t confirm this because the deal is not yet closed. It doesn’t mean that CNL didn’t overpay, but it helps to explain how they could justify the pricing. Kudos should probably go to Thilo Best, Horizon Bay’s CEO, for positioning the company’s assets to be sold at such a high price.
In addition to cap rates, industry participants have questioned the credit worthiness of CNL’s tenants. It is true that many of the lessees are thinly capitalized, but in CNL’s mind the tenant is only as good as the properties, and since the properties are what provide the cash flow anyway, the financial strength, or lack thereof, of the shell lessee loses its importance. The point is understood, and makes sense in theory, but it also remains a debatable, and controversial, issue.
CNL, with its seemingly unlimited supply of equity, fueling an acquisition appetite never seen before in the senior care industry, may be a temporary market phenomenon because so many small investors are looking for yield. Or it may help in getting the attention of other potentially large providers of equity capital that are eyeing the senior care market. Or, it may be indicative of a change in the acquisition market, especially how portfolios are valued. If my cost of capital is, for example, 2%, with an investment horizon of 20 years, why should I be bound by a "market" cap rate of 10%? In theory, this is true, and the industry went through this in the 1980s with the phony nonprofit entities and their low cost of capital, with often disastrous results. The problem is, just because your cost of capital is low, it does not mean you should pay up (even though you can).
This brings up what may become the most controversial development in the market this year, and that is how changes in capital flow will impact the acquisition market. With pension funds such as CalPERS, and now CNL, taking the leading role in the market, and paying prices that operators would not pay, there is the potential of producing a divided market. In the case of the Horizon Bay transaction, no operating company would have come close to paying CNL’s price, because it wouldn’t have the necessary equity, lenders would not have financed it at that level and they would not pay for the fill-up risk.
Last year, Horizon Bay refinanced 10 properties, representing almost 46% of its units, for just $76,800 per unit, or less than half of what CNL paid on a per-unit basis for the entire portfolio. While this may or may not be relevant, the point is that companies such as Sunrise and Brookdale Living Communities, to name two, would not touch the price of $562 million for Horizon Bay (and Goldman knew it), but they would manage the properties for someone else. So now, every time someone wants to sell, the first option will be the "financial buyer," or real estate investor, and then an operating company. If this trend continues, it will have ramifications for the ability of regional, and even national, companies to grow other than by management contracts or one-off acquisitions.
The second point is that as the seniors housing market continues to stabilize, grow and increase its profits, more institutional equity will be providing fresh capital, especially if alternative rates of return remain so low. What they bring to the market is a lower rate of return expectation, and they are attracted to the seniors housing market in part because they see opportunity where there is such a large spread between seniors housing cap rates and cap rates for other types of real estate. The spread is nothing particularly new, but the increasing acceptability of seniors housing as an asset class among traditional institutional equity investors is.
Consequently, unless there is another industry blow-up on the horizon, there very well could be a fundamental downward shift in cap rates, for both independent living communities and higher-end assisted living facilities, but with the impact being felt first with the former property type. Single digit cap rates will become more common, as opposed to currently being the exception to the rule, and asset "values" will rise. There is always the risk that this new atmosphere in the market will contribute to its own undoing, and given the cyclical nature of business and interest rates, there will certainly be another downturn in the senior care market at some point. But the future is looking pretty good right now, and that is attracting new equity in search of diversification, and yield.
Now, we would be remiss if we did not bring up a few caveats. First of all, other real estate asset classes are still viewed as pure real estate plays, compared with seniors housing which, to varying degrees, has a significant operational aspect. And while a class B office building is a class B office building, the same does not hold true for assisted living, for example. Second, there is still an absence of total return data for the senior care sector, since no one has really exited the market after a period of time and reported any kind of average annual return. That is, no one who has had good news to report, with the exception now of Goldman Sachs, but we may never know what its final return on invested capital really was. Third, if interest rates shoot up by 200 to 400 basis points, this would change the entire scenario, especially where rates of return fit in among the various investment alternatives.
Finally, there is a difference between the capital coming into the senior care market today compared with the 1990s. In the past decade, most of the money was fueling development, and growth was as close to 100% leveraged as you can get. Today, most of the capital is going into existing properties, which by definition should be less risky, and the leverage is much lower. Even if you take the view that CNL, for example, is overpaying by 25%, its leverage today would be just 33%. Despite this, we still believe the price is too rich for our blood. But who knows, three years from now we may be looking at some of CNL’s deals and talking about what a deal they were. Let’s hope so.
The astute reader (who is still awake) may be wondering why the skilled nursing sector has been left out of this discussion. The problem is that with so much of their revenues coming from the government, and as an asset class where operations still hold supreme, most "real estate" investors will continue to take a pass and return expectations will continue to be high. Even though private equity is creeping its way back into the SNF market, we do not expect to see cap rates decline much until some of the reimbursement and litigation issues are worked out on a national basis.